ultrasound signal processing circuitry and related apparatus and methods are described. signal samples received from an ultrasound transducer array in an ultrasound transducer based imaging system may be processed, or conditioned, by application of one or more weighting functions. In some embodiments, one or more weighting functions may be applied to the signal samples in the time domain. In other embodiments, the signal samples may be converted to the frequency domain and one or more weighting functions may be applied in the frequency domain. In further embodiments, one or more weighting functions may be applied in the time domain and one or more weighting functions may be applied in the frequency domain. The weighting functions may be channel dependent and/or channel independent. The processed data can be provided to an image formation processor.
|
1. An ultrasound device comprising:
an ultrasound transducer array configured to provide received signals in response to receiving ultrasound energy; and
a processing circuit configured to process the received signals, the processing circuit comprising:
a conversion circuit configured to convert the received signals to a digital domain to provide signal samples;
an extraction circuit configured to extract from the signal samples a subset of the signal samples that correspond to an image to be formed;
a time domain signal conditioning circuit configured to apply at least one time domain weighting function to the extracted signal samples;
a Fast fourier Transform circuit configured to convert the weighted signal samples to frequency domain values;
a frequency domain signal conditioning circuit configured to apply at least one frequency domain weighting function to the frequency domain values; and
an output circuit configured to output the weighted frequency domain values for ultrasound image formation processing;
wherein the processing circuit is configured to partially process the extracted signal samples in a first number of channels, to store the partially processed signal samples in a memory, and to complete processing of the partially processed signal samples in a second number of channels less than the first number of channels.
2. The ultrasound device of
3. The ultrasound device of
4. The ultrasound device of
5. The ultrasound device of
6. The ultrasound device of
|
This application is a Continuation claiming the benefit under 35 U.S.C. § 120 of U.S. application Ser. No. 15/517,284, filed Apr. 6, 2017, and entitled “ULTRASOUND SIGNAL PROCESSING CIRCUITRY AND RELATED APPARATUS AND METHODS,” which is hereby incorporated herein by reference in its entirety.
U.S. application Ser. No. 15/517,284 is a National Stage Application of PCT/US2015/054405, filed Oct. 7, 2015, and entitled “ULTRASOUND SIGNAL PROCESSING CIRCUITRY AND RELATED APPARATUS AND METHODS,” which is hereby incorporated herein by reference in its entirety.
Patent Application Serial No. PCT/US2015/054405 claims the benefit under 35 U.S.C. § 119(e) of U.S. Provisional Patent Application Ser. No. 62/060,822, filed Oct. 7, 2014 and entitled “ULTRASOUND SIGNAL PROCESSING CIRCUITRY AND RELATED APPARATUS AND METHODS,” which is hereby incorporated herein by reference in its entirety.
Aspects of the present disclosure relate to circuitry, devices, systems and methods for imaging and/or treatment, such as ultrasonic imaging and/or treatment technology. More particularly, aspects of the present disclosure relate to circuitry and methods for processing signals received from an ultrasound transducer array.
Ultrasound transducer arrays used for medical applications typically produce a large amount of data, as needed to produce ultrasound images for medical applications. The higher the quality and complexity of the desired images, the more data is typically needed.
The problem of transporting multiple channels of analog signals from an ultrasound transducer array to the control and processing electronics of an ultrasound system has limited the utility of the larger and denser arrays of transducers needed to improve the resolution of ultrasound imaging and to enable high quality 3D volumetric imaging.
The present disclosure describes aspects of processing signals received from an ultrasound transducer array in an ultrasound transducer based imaging system, including the digital and analog circuitry used to process the signals. In some embodiments, signal samples are processed, or conditioned, by application of one or more weighting functions. In some embodiments, one or more weighting functions may be applied to the signal samples in the time domain. In other embodiments, the signal samples may be converted to the frequency domain and one or more weighting functions may be applied in the frequency domain. In further embodiments, one or more weighting functions may be applied in the time domain and one or more weighting functions may be applied in the frequency domain. The weighting functions may be channel dependent and/or channel independent. The processed data can be provided to an image formation processor. The processing of signals prior to image formation processing may be termed “preprocessing” of the signals received from the ultrasound transducer array.
Some embodiments are directed to a method for processing signals received from an ultrasound transducer array. The method comprises signal conditioning of the received signals after conversion of the received signals from an analog domain to a digital domain.
Some embodiments are directed to a method for processing signals received from an ultrasound transducer array. The method comprises converting the received signals to a digital domain to provide signal samples and performing quadrature demodulation of the signal samples followed by Fast Fourier Transform of the demodulated signal samples and signal conditioning in a frequency domain.
Some embodiments are directed to a method for processing signals received from an ultrasound transducer array. The method comprises summing elevation channels of the ultrasound transducer array in a frequency domain following Fast Fourier Transform of the received signals.
Some embodiments are directed to a method for processing signals received from an ultrasound transducer array. The method comprises processing the received signals to provide signals for Fourier Resample image formation processing and/or Back Projection image formation processing.
Some embodiments are directed to a method for processing signals received from an ultrasound transducer array. The method comprises processing the received signals with a first number of channels to provide partially processed signal samples, storing the partially processed signal samples in a memory, and completing processing of the partially processed signal samples with a second number of channels less than the first number of channels.
Some embodiments are directed to a method for processing signals received from an ultrasound transducer array. The method comprises converting the received signals to a digital domain to provide signal samples, conditioning the signal samples, and outputting the conditioned signal samples for image formation processing.
Some embodiments are directed to an ultrasound device comprising: an ultrasound transducer array configured to provide received signals in response to receiving ultrasound energy, and a processing circuit configured to process the received signals. The processing circuit comprises a conversion circuit configured to convert the received signals to a digital domain to provide signal samples, a conditioning circuit configured to condition the signal samples, and an output circuit configured to output the conditioned signal samples for image formation processing.
Some embodiments are directed to a method for processing signals received from an ultrasound transducer array. The method comprises converting the received signals to a digital domain to provide signal samples, extracting from the signal samples a subset of the signal samples that correspond to an image to be formed, application of a time domain weighting function to the signal samples, conversion of the weighted signal samples to frequency domain values, application of a frequency domain weighting function to the frequency domain values, and outputting the weighted frequency domain values for image formation processing.
Various aspects and embodiments of the disclosed technology will be described with reference to the following figures. It should be appreciated that the figures are not necessarily drawn to scale. Items appearing in multiple figures are indicated by the same reference number in all the figures in which they appear.
Aspects of the present disclosure relate to digital and analog circuitry and methods for processing signals received from an ultrasound transducer array. In some embodiments, the ultrasound transducer array and the circuitry may be integrated on a single complementary metal oxide semiconductor (CMOS) chip, or substrate, or may be on multiple chips within an ultrasound probe. The present disclosure provides unique, cost-effective, and scalable integrated signal processing architectures to process signals from ultrasound transducer elements or groups of ultrasound transducer elements and to provide data that is sufficiently robust for advanced high quality imaging applications. Thus, aspects of the present disclosure provide an architecture which may be used with a single substrate ultrasound device having integrated ultrasound transducers (e.g. CMOS ultrasonic transducers) and digital circuitry.
The present disclosure describes aspects of processing signals received from an ultrasound transducer array in an ultrasound transducer-based imaging system. In some embodiments, signal samples are processed, or conditioned, by application of one or more weighting functions, or masks. In some embodiments, one or more weighting functions may be applied to the signal samples in the time domain. In other embodiments, the signal samples may be converted to the frequency domain and one or more weighting functions may be applied in the frequency domain. In further embodiments, one or more weighting functions may be applied in the time domain and one or more weighting functions may be applied in the frequency domain. The weighting functions may be channel dependent and/or channel independent. The processed data can be provided to an image formation processor. The processing of signals prior to image formation processing may be termed “preprocessing” of the signals received from the ultrasound transducer array.
In addition, the signal samples may be converted to a form that is advantageous for image formation processing. For example, data corresponding to several elevation channels can be combined prior to image formation processing. In general, various signal processing functions may be performed prior to image formation processing or during image formation processing based on a particular architecture. The signal processing architecture may further include data reduction, compression and/or downsampling to reduce the volume of data being processed. Such operations may include, for example, quadrature demodulation, filtering and downsampling. In further embodiments, signal samples that do not contribute to the image being formed or which degrade the image may be discarded.
The aspects and embodiments described above, as well as additional aspects and embodiments, are described further below. These aspects and/or embodiments may be used individually, all together, or in any combination of two or more, as the application is not limited in this respect.
It should be appreciated that communication between one or more of the illustrated components may be performed in any of numerous ways. In some embodiments, for example, one or more high-speed busses (not shown), such as that employed by a unified Northbridge, may be used to allow high-speed intra-chip communication or communication with one or more off-chip components.
The one or more transducer arrays 102 may take on any of numerous forms, and aspects of the present technology do not necessarily require the use of any particular type or arrangement of transducer cells or transducer elements. Indeed, although the term “array” is used in this description, it should be appreciated that in some embodiments the transducer elements may not be organized in an array and may instead be arranged in some non-array fashion. In various embodiments, each of the transducer elements in the array 102 may, for example, include one or more CMUTs, one or more CMOS ultrasonic transducers (CUTs), and/or one or more other suitable ultrasonic transducer cells. In some embodiments, the transducer elements of the transducer array 102 may be formed on the same chip as the electronics of the TX circuitry 104 and/or RX circuitry 106. Numerous examples of ultrasonic transducer cells, elements, and arrangements (e.g., arrays), as well as methods of integrating such devices with underlying CMOS circuitry, are discussed in detail in U.S. patent application Ser. No. 14/208,351, entitled COMPLEMENTARY METAL OXIDE SEMICONDUCTOR (CMOS) ULTRASONIC TRANSDUCERS AND METHODS FOR FORMING THE SAME, and filed on Mar. 13, 2014, the entire disclosure of which is incorporated herein by reference.
A CUT may, for example, include a cavity formed in a CMOS wafer, with a membrane overlying the cavity, and in some embodiments sealing the cavity. Electrodes may be provided to create a transducer cell from the covered cavity structure. The CMOS wafer may include integrated circuitry to which the transducer cell may be connected. The transducer cell and CMOS wafer may be monolithically integrated, thus forming an integrated ultrasonic transducer cell and integrated circuit on a single substrate (the CMOS wafer).
The TX circuitry 104 (if included) may, for example, generate pulses that drive the individual elements of, or one or more groups of elements within, the transducer array(s) 102 so as to generate acoustic signals to be used for imaging. The RX circuitry 106, on the other hand, may receive and process electronic signals generated by the individual elements of the transducer array(s) 102 when acoustic signals impinge upon such elements.
In some embodiments, the timing & control circuit 108 may, for example, be responsible for generating all timing and control signals that are used to synchronize and coordinate the operation of the other elements in the device 100. In the example shown, the timing & control circuit 108 is driven by a single clock signal CLK supplied to an input port 116. The clock signal CLK may, for example, be a high-frequency clock used to drive one or more of the on-chip circuit components. In some embodiments, the clock signal CLK may, for example, be a 1.5625 GHz or 2.5 GHz clock used to drive a high-speed serial output device (not shown in
The power management circuit 118 may, for example, be responsible for converting one or more input voltages VIN from an off-chip source into voltages needed to carry out operation of the chip, and for otherwise managing power consumption within the device 100. In some embodiments, for example, a single voltage (e.g., 12V, 80V, 100V, 120V, etc.) may be supplied to the chip and the power management circuit 118 may step that voltage up or down, as necessary, using a charge pump circuit or via some other DC-to-DC voltage conversion mechanism. In other embodiments, multiple different voltages may be supplied separately to the power management circuit 118 for processing and/or distribution to the other on-chip components.
As shown in
Moreover, it should be appreciated that the HIFU controller 120 may not represent distinct circuitry in those embodiments providing HIFU functionality. For example, in some embodiments, the remaining circuitry of
In addition to using different power levels, imaging and HIFU applications may utilize different waveforms. Thus, waveform generation circuitry may be used to provide suitable waveforms for operating the system as either an imaging system or a HIFU system.
In some embodiments, the system may operate as both an imaging system and a HIFU system (e.g., capable of providing image-guided HIFU). In some such embodiments, the same on-chip circuitry may be utilized to provide both functions, with suitable timing sequences used to control the operation between the two modalities. Additional details with respect to HIFU implementations and operational features that may be employed in the various embodiments set forth in the present disclosure are described in U.S. patent application Ser. No. 13/654,337, entitled TRANSMISSIVE IMAGING AND RELATED APPARATUS AND METHODS, filed Oct. 17, 2012, the entire contents of which is incorporated herein by reference.
In the example shown, one or more output ports 114 may output a high-speed serial data stream generated by one or more components of the signal conditioning/processing circuit 110. Such data streams may, for example, be generated by one or more USB 3.0 modules, and/or one or more 10 GB, 40 GB, or 100 GB Ethernet modules, integrated on the die 112. In some embodiments, the signal stream produced on output port 114 can be fed to a computer, tablet, or smartphone for the generation and/or display of 2-dimensional, 3-dimensional, and/or tomographic images. In embodiments in which image formation capabilities are incorporated in the signal conditioning/processing circuit 110, even relatively low-power devices, such as smartphones or tablets which have only a limited amount of processing power and memory available for application execution, can display images using only a serial data stream from the output port 114. As noted above, the use of on-chip analog-to-digital conversion and a high-speed serial data link to offload a digital data stream is one of the features that helps facilitate an “ultrasound on a chip” solution according to some embodiments of the present disclosure.
Devices 100 such as that shown in
In yet other implementations, a pair of imaging devices can be positioned so as to straddle a subject, such that one or more CMUT elements in the device(s) 100 of the imaging device on one side of the subject can sense acoustic signals generated by one or more CMUT elements in the device(s) 100 of the imaging device on the other side of the subject, to the extent that such pulses were not substantially attenuated by the subject. Moreover, in some implementations, the same device 100 can be used to measure both the scattering of acoustic signals from one or more of its own CMUT elements as well as the transmission of acoustic signals from one or more of the CMUT elements disposed in an imaging device on the opposite side of the subject.
In the example shown in
As shown in
As shown, the TX circuitry 104 for a respective transducer element 204 may include both a waveform generator 206 and a pulser 208. The waveform generator 206 may, for example, be responsible for generating a waveform that is to be applied to the pulser 208, so as to cause the pulser 208 to output a driving signal to the transducer element 204 corresponding to the generated waveform.
In the example shown in
After undergoing processing in the digital processing block 214, the outputs of all of the RX circuits on the die 112 (the number of which, in this example, is equal to the number of transducer elements 204 on the chip) are fed to a multiplexer (MUX) 216 in the signal conditioning/processing circuit 110. In other embodiments, the number of transducer elements is larger than the number of RX circuits, and several transducer elements provide signals to a single RX circuit. The MUX 216 multiplexes the digital data from the RX circuits, and the output of the MUX 216 is fed to a multiplexed digital processing block 218 in the signal conditioning/processing circuit 110, for final processing before the data is output from the die 112, e.g., via one or more high-speed serial output ports 114. The MUX 216 is optional, and in some embodiments parallel signal processing is performed. A high-speed serial data port may be provided at any interface between or within blocks, any interface between chips and/or any interface to a host. Various components in the analog processing block 210 and/or the digital processing block 214 may reduce the amount of data that needs to be output from the die 112 via a high-speed serial data link or otherwise. In some embodiments, for example, one or more components in the analog processing block 210 and/or the digital processing block 214 may thus serve to allow the RX circuitry 106 to receive transmitted and/or scattered ultrasound pressure waves with an improved signal-to-noise ratio (SNR) and in a manner compatible with a diversity of waveforms. The inclusion of such elements may thus further facilitate and/or enhance the disclosed “ultrasound-on-a-chip” solution in some embodiments.
Although particular components that may optionally be included in the analog processing block 210 are described below, it should be appreciated that digital counterparts to such analog components may additionally or alternatively be employed in the digital processing block 214. The converse is also true. That is, although particular components that may optionally be included in the digital processing block 214 are described below, it should be appreciated that analog counterparts to such digital components may additionally or alternatively be employed in the analog processing block 210.
As shown in
In the example of
In some embodiments, it may be desirable to match the center frequency “fc” of the DQDM 308 with the frequency of interest of the transducer elements that are used in the array(s) 102. Examples of additional components that may, in some embodiments, be included in RX circuitry 106, in addition to or in lieu of the DQDM 308 and/or the other components illustrated in
An illustrative embodiment of a circuit suitable for use as the matched filter 402 is shown in
To operate as a “matched” filter, the value of “H(ω)” applied to the multiplier 506 should be a conjugate of the transmission waveform Tx(ω). In some embodiments, the filter 402 may thus indeed operate as a “matched” filter, by applying a conjugate of the transmission waveform Tx(ω) to the multiplier 506. In other embodiments, however, the “matched” filter 402 may instead operate as a mismatched filter, in which case some value other than a conjugate of the transmission waveform Tx(ω) may be applied to the multiplier 506.
An example of a digital dechirp circuit 602 is shown in
Digital Signal Processing Circuitry
As shown in
The signal processing circuit of
The signal processing circuit of
Extract Range Swath Block
The extract range swath block 910 selects input samples that contribute to the image and discards input samples that do not contribute to the image. In order to process an image whose pixels have a given extent and location relative to the aperture, and a waveform with a given pulse length is used, there is a set of time samples that will contribute to the image pixels for a given receiver/excitation combination; time samples outside this set can be discarded. In some embodiments, the extract range swath block 910 may be implemented by streaming of data from the ADC 212, wherein the selected range of data is defined by the beginning and ending times when the data is digitized and/or is injected into the signal processing circuit.
Extracting the contributing portion of the receive swath can reduce the data transfer requirements (when done on-board), the data storage requirements (whether in memory or writing to disk), and the processing burden. This can be done to various degrees of compactness depending on the importance of the data reduction. A basic implementation includes a constant time extent across all receivers and all excitations, with a constant start time across all receivers and all excitations. Other implementations can use a separate start time and time extent for each receiver and each excitation. After data transfer, the data are aligned and arranged in whatever form is necessary for processing.
There are usually nonzero receive A/D samples at times while the system is transmitting or shortly thereafter, resulting in highly distorted A/D values from saturation or other nonlinearities, despite any receiver protector circuitry or switching. These samples do not contribute to usable imagery and can cause many problems and artifacts in the imagery, which make it generally more difficult to do basic diagnostics. When performing any sort of deconvolution or other temporal frequency domain processing (often even just truncating to a processing band), the energy in the extended time domain may contaminate the entire image. Making estimates of the spectrum (either for diagnostics or calibration) with these samples present can be problematic, since the energy in these samples dominates the energy in the entire receive channel.
These samples may be discarded during preprocessing. The approximate index where this nonlinear portion ends can be determined using the relative delay information and the pulse length of the waveform. An additional buffer can be used to be sure that the nonlinear samples are all identified. This process may be performed independently across channel and excitation to minimize the amount of the image that is discarded at near range.
The data at the input to the preprocessor may be real or complex, and may already have an implied carrier frequency. The combined steps of carrier adjustment, low pass filtering, and downsampling ensure that the data are complex, not excessively oversampled, and have the desired carrier frequency for the image formation processor. The existing carrier frequency may be a “default” one and may not be the actual center of the desired processing band.
Data Reduction
In some embodiments of the present disclosure, a cascade integrating comb (CIC) filter architecture may be used to perform filtering (e.g., for filter block 914) and decimation (e.g., for downsample block 916). For example, such a CIC filter architecture may be used to accurately calculate a range value using a precise delay time index. The CIC filter includes a plurality (N) stages and acts as a low-pass filter, while decimating the input data stream x[n] to produce an output data stream y[n]. Increasing the number of stages results in more droop in the passband, while increasing the number of stages results in better image rejection. In some implementations, passband droop may be at least partially addressed using a compensation filter that is applied after the CIC filter has been applied to the data.
The circuit of
As shown in
In the illustrative architecture of
The outputs of downsample block 916 are passed to the comb stage 1012b of the CIC filter. As shown, comb stage 1012b includes delay elements 1050 and subtraction elements 1052. The outputs of the comb stage 1012b are passed to re-quantization circuit 1016, where re-quantization of the digital signals is performed using re-quantization circuits 1060. The outputs of re-quantization circuit 1016 are passed to arithmetic logic unit (ALU) 1018, which provides additional arithmetic processing.
Memory
Referring again to
Memory can be provided between any pair of blocks or even sub-blocks (blocks within blocks). At any point in the processing circuit, a memory block may facilitate a reduction in the rate of streamed processing, thus reducing the number of parallel resources needed for processing, e.g., 1152 channels being processed concurrently may be saved to memory, then after memory the streaming processing may only consist of 4 channels at a time. One reason for reducing the streaming rate is to optimize between speed and resources by matching a data rate interface, e.g., universal serial bus (USB), Firewire, low voltage differential signaling (LVDS), Thunderbolt, or other.
Time Domain Signal Conditioning
The time domain signal conditioning block 922 shown in
In the embodiment of
The received signal may need to be altered across time and/or range in order to produce imagery with desired characteristics. This may be done using weightings in the time or range compressed domain. Weightings performed in almost any domain may be done to account for physically relevant phenomena. Examples are time invariant transfer functions applied as weights in the frequency domain, time-dependent weights to account for TGC (time-gain compensation), and range-dependent weights to account for attenuation/“range loss”. The time domain is to be distinguished from the range compressed domain. Weights applied across time and weights applied across range mean different things when there are sufficiently long waveforms imposed on the time domain data. There are effects that are more accurately described as time domain effects, such as TGC or other time-dependent receiver gains, and effects that are more accurately described as range domain effects, such as tissue attenuation.
An accurate preprocessor (or forward operator, depending on which mode the processing is being used/defined) separates the application/removal of time and range domain weights, with waveform and system transfer function application/removal occurring between the two. Sharp transitions in range and/or time need to be applied with care, since these mean physically different things when extended waveforms are present, and since steep slopes and transitions affect the shape of the spectrum (relevant when deconvolving using the data itself; a ramp in the range/time domain is a derivative in the temporal spectrum). When the situation, parameters, or desired image quality dictate that the time and range weights are to be applied separately, true range processing can be used.
In order to perform “true range weighting” separately from time domain weighting, additional FFTs may be needed depending on the form of the remainder of the preprocessing chain and the definition of the output preprocessed data domain. There are many ways to describe all of the potential combinations for doing this. One of the most computationally efficient preprocessing combines the fast-time and range weightings into a single set of weights that are applied along time. When the weights are combined, the range-dependent weights are moved to the time domain.
Note that if FFT shifting the output is desired after the FFT (preprocessing block after the time domain weightings), that can be accomplished by a multiplication along time before the FFT. The linear phase ramp to accomplish this can be absorbed into the precomputed time domain weighting for no additional computational cost during preprocessing.
Whether the time domain weighting is channel-dependent or channel-independent, it may be desirable to perform dispersive (frequency dependent) time domain weighting (or “true range weighting”). This can be done several ways, including by polynomial or other basis expansion, and multirate filter banks.
The most basic case of time domain weighting is a channel-independent (receiver and excitation independent) weighting. When the only weighting to be applied across time is channel-independent, then there can be a savings of memory and a simplification in the indexing. When any other form of time domain weighting (receiver-dependent, excitation-dependent, or channel-dependent) is used, this channel-independent weight can be absorbed into the other time domain weighting. Examples of channel-independent time domain weights include: (1) carrier frequency adjustment, as discussed above; (2) linear phase that when applied in time, FFT shifts the temporal frequency domain after the FFT; and (3) time-gain compensation (TGC) profile, which in some cases is the same for every receiver and for every excitation.
There may be receiver-dependent fast-time domain weights that are a function of time but not a function of excitation. An example is TGC profiles that are different enough from receiver-to-receiver that they need to be compensated for individually (this may still be the case even if the TGC settings/parameters are the same for every receiver; this is possible if the variation in the amplifier gains are so great that they need to be dealt with separately).
There may be a need to apply excitation-dependent fast-time domain weights. This could be the case when the excitations are different enough that different TGC settings are intentionally used to best quantize the signal across all excitations.
When the scenario is such that the range-dependent weights are being combined with the time domain weights, if there is excitation-dependent excess time delay, then the time/range lineup changes from excitation-to-excitation. One way to maintain the same time/range relationship across excitation is to have the same delay at the center (FFT center, not the mean) of the aperture.
Channel-dependent time domain weights may be applied within a preprocessing chain. It is also possible to have a set of receiver-dependent time domain weights and a set of excitation-dependent time domain weights, and this creates a trade off between the memory storage of combining these into a single set of channel-dependent weights vs. two separate sets of multiplies with more complicated indexing.
The primary true time dependent weighting of relevance is Time-Gain Compensation (TGC). Specific range dependent weights that might be absorbed/combined with the time domain weights are discussed below. The TGC profile and its correction weighting (usually the reciprocal of the profile, possibly with some regularization) should be provided as narrowband information from each specific sensor.
Fast Fourier Transform Block
A schematic block diagram of an example of the FFT block 924 of
In some embodiments, the FFT units 1210 may be 1024 point variable streaming FFT units. However, other FFT units may be utilized. In the example of
The FFT block 924 of
One factor is the length of the time domain waveform and all other system time domain impulse responses being convolved onto/removed from the data via “matched filtering” (true matched filters, have the same convolution length as the original signal). This is relevant when using preprocessing chains in the forward scattering and the adjoint of forward scattering modes.
Another factor is the length of the time domain mismatched filter when processing data in the “inverse” and the adjoint of “inverse” modes. The term “mismatched filter” here means a reference signal that is not the matched filter. The mismatched filter generation technique that is of particular use is simply the signal generated by regularized deconvolution of the signal's spectrum. Poor choices of mismatched filters will not adequately remove the waveform, while useful mismatched filters likely have effective lengths equal to or longer than matched filters. In some cases, an extended length may be used to move unwanted signals far out such that they can easily be removed. There may be a large enough difference between the lengths of the signals being convolved for forward/adjoint processing than for “inverse”/adjoint of “inverse” processing that separate chains may be more efficient and more appropriate.
When the length of the mismatched filter is intentionally made to be very long, it may be appropriate to add an additional pair of FFTs and pads/truncations/weights in range to remove the unwanted signal pushed by the mismatched filter as well as to equate the temporal frequency grids (number of frequencies, spacing, and start frequency) between all modes for the final preprocessed complex scattering function.
While a single branch that also uses these same FFT sizes and additional FFT pairs for the forward and adjoint modes is possible, it is not enforced that the same FFT sizes and additional FFT pairs be used for the “inverse.” Using a “good FFT size” makes a big difference in the speed of the FFT, so the next larger “good FFT size” is typically used after taking into account all other sizing information.
One effect of the FFT is a shift in represented frequencies. Often this remapping is considered a quadrant swap/“FFTshift” used when data is going into and out of a DFT (Discrete Fourier Transform)/FFT. Solely moving memory with FFTshifting may often incur a latency penalty, but algorithms can be “streamlined” to remove FFTshift penalties. The FFT shifts can be performed while inserting the nonzero data into a zeropadded array or when pulling the data from the 1-D buffer where the processing takes place.
Another way to perform FFTshifts indirectly is by multiplying by linear phase ramps (a linear phase ramp multiplied on the input to the FFT can result in an FFTshift of the output, a linear phase ramp multiplied on the output of the FFT can result in the same effect as applying an FFTshift to the input). The linear phase ramps that perform this are simple (±1 on both sides when length is divisible by four, ±1 or ∓1 on the different sides when length is only divisible by two. When one considers that there are not many genuine “good FFT sizes” of appreciable length that are not evenly divisible by four, then these linear phase ramps are just ±1 across the fast-time or temporal frequency samples on both sides of the FFT.
By combining these linear phase ramps with the time domain multiplies and the frequency domain multiplies, FFTshifting in and out of the FFT is performed at no additional computational burden after the initial computation of the combined weights.
After the FFT, the spectrum is trimmed to the portion of the processing band that is chosen to contribute to the image. This may be chosen from image quality requirements.
Frequency Domain Signal Conditioning
A schematic block diagram of an example of the frequency domain signal conditioning block 926 of
Referring to
The RAM 1320 may be sized to accommodate channel independent or channel dependent multiplication. Furthermore, an implementation may be dependent on the receiver transducer location, e.g., array row channel dependent and column channel independent or vice versa. Here the NCO is generating a single frequency for multiplication, which has the effect of imparting a delay in the associated time domain signal (the signal after inverse FFT). The first multiply is a method for imparting a delay and an apodization.
The frequency domain preprocessing and weighting performs the bulk of the processing on the data and has the biggest impact on image quality. This is where the transfer functions of all the individual pieces are combined and accounted for, and where motion compensation/phase adjustment can be performed.
If FFT shifting the input is desired before the fast-time FFT (preprocessing block preceding the temporal frequency domain weightings), that can be accomplished by a multiplication along temporal frequency after the FFT. The linear phase ramp to accomplish this can be absorbed into any of the precomputed temporal frequency weightings for no additional computational cost during preprocessing.
There are many options and combinations for combining channel-independent and receiver/excitation/channel-dependent weightings. The basic forms are discussed here, and specific choices are left to the specific scenarios.
A channel-independent frequency domain weighting may be desirable to account for several effects: (1) temporal frequency linear aperture weighting, chosen to impose a specific sidelobe structure in the imagery, rather than just using whatever the system generated with its non-flat waveform generator/transfer function combination; (2) constant “master waveform” applied across all channels; and (3) common transducer transfer function.
In most cases, there is at least one receiver/excitation/channel-dependent frequency domain weighting that needs to be applied within preprocessing, and the channel-independent frequency domain weighting can be absorbed there. One possible exception may be when the receiver/excitation/channel-dependent frequency domain weighting is phase-only, where the phase is described by a low-order polynomial (such as motion compensations with linear phase or other phase adjustments with quadratic phase functions). In this case, the phase-only function can be efficiently computed on-the-fly, and the channel-independent weight is applied as a separate multiply step. This would incur more overall multiplies but save a large amount of memory that would be used for storing the pre-computed weights (particularly for fully channel-dependent weights).
Receiver-dependent frequency domain weighting may be useful. This would be the case if the transfer functions of each combined transmitter/transducer/receiver are different enough to warrant accounting for them separately.
Excitation-dependent frequency domain weighting may be applied to the data that is receiver-independent. A relevant example is for plane wave excitations, where there is often an offset delay relative to the phase reference at the middle receiver that is a function of the plane wave angle. While this can be absorbed into a time domain interpolation or a fully channel-dependent frequency domain weighting, the amount of memory storage for a full set of weights may make an excitation-dependent weighting attractive.
Channel-dependent frequency domain weighting may also be utilized. The most general weighting is one that is potentially different on every channel of data, where a channel is a unique receiver/excitation combination. Any channel-independent weightings can be absorbed into channel-dependent, receiver-dependent, or excitation-dependent weightings.
When using a receiver-dependent weighting and an excitation-dependent weighting, there may a tradeoff between the additional storage required to absorb both weightings into a single channel-dependent weighting vs. using less storage with two separate multiplies.
There may be a need to provide frequency-independent, time/range independent, but channel-dependent weightings. The most common type of this weighting is a scalar gain that is different from receiver-to-receiver, but is constant across excitation. These weights would likely have the fewest number of coefficients (since fast-time A/D samples dominate over the number of receivers), but if every complex multiply is expensive, then these types of weights can be absorbed into the other channel-dependent weights in whatever way is most appropriate (in fast-time or along frequency, depending on which one has a corresponding weight set that has the same receiver/excitation/channel dependence). If there are no receiver/excitation/channel-dependent corrections anywhere within preprocessing, then the tradeoff could be made between having a separate multiply stage vs. storage of fully channel-dependent weights where these fast-time/frequency-independent weights can be absorbed.
True range processing may be performed separately from other weighting. There may be scenarios, particularly with long waveforms, where it may be desirable to apply weights before and after waveform application/removal to better emulate fast-time and range domain physical processes. The choice to separate these vs. lump them into the fast-time weights is dependent on the specific scenario and constraints.
Reduction of localized acoustic energy as it propagates through tissue can be significant. It may be desirable to unweight the raw data with an estimated range-dependent profile in order to level out the image. It may be useful to compensate for approximate range decay. In particular, many 2-D imaging formulations assume infinite line sources and infinite line transducer elements, which result in cylindrical wave decay. Many of these formulations actually impose the correct cylindrical wave behavior on the raw data (when used in the forward sense, and accurately removed when used in the “inverse” sense). But since the actual transducers behave more like point sources and the volume is composed of point scatterers, spherical waves are more appropriate to describe basic propagation loss.
The signal attenuation characteristics through tissue are not generally known in advance. Approximating the attenuation as a homogeneous process with estimated parameters, however, can aid in leveling out the image brightness as a function of downrange. Even with assumed homogeneous attenuation parameters, the attenuation should be imposed/removed as a function of frequency, either through polynomial or other basis expansion, multirate, or by other means. If this is too computationally burdensome, then it can be approximated using the parameters at a single frequency.
Sum Elevation Channels
A schematic block diagram of an example of the sum elevation channels block 930 of
Referring to
In some scenarios, after the frequency domain weighting, it may be necessary to filter and resample the data along temporal frequency. This can be done by low pass filtering/downsampling/resampling, or by pairs of FFTs with another weighting in the middle. This generic form may be kept in mind when considering the various processing options.
Data in the range compressed domain may be the required output of the preprocessor/input to image formation, e.g. Backprojection. Regardless of domain choice, standardized data ports may be established along the processing chain, whether or not that port is explicitly a waypoint along a given preprocessor/IFP processing chain. To get to the range compressed data port, another FFT (IFFT, if an FFT was done along fast-time) is performed after the frequency domain weighting. The IFFT 932 of
As mentioned, the output of the preprocessor is data that is to be ingested by an image formation processor (IFP). Preferably, this input to the image formation processor should be in the form of complex scattering function in the temporal frequency vs. receiver channel domain, but this is not the only option. Branches that have the input of the image formation processor in the range compressed domain are useful for reconstructions such as backprojection.
Channel Configuration
A schematic block diagram of an example of a channel configuration of the signal processing architecture is shown in
In the example of
Digital Signal Processing Methods
A flowchart that illustrates an example of a method performed by the signal processing circuit of
In stage 1630, data values are read from memory 920 and time domain signal conditioning is performed by time domain signal conditioning block 922. As described above, time domain signal conditioning may include application of one or more weighting functions to the time domain signal. In stage 1632, a Fast Fourier Transform is applied to the signal samples, and frequency domain signal conditioning is performed in stage 1634. As described above, frequency domain signal conditioning involves application of one or more frequency domain weighting functions to the frequency domain data. In stage 1636, the elevation channels are summed by sum elevation channels block 930 to thereby reduce the quantity of data supplied for image formation processing. In stage 1638, an inverse Fast Fourier Transform may be applied to the conditioned signal samples if time domain signals are required for image formation processing. In stage 1640, the conditioned signal samples are utilized for image formation processing.
In the process of
Having thus described several aspects and embodiments of the technology set forth in the disclosure, it is to be appreciated that various alterations, modifications, and improvements will readily occur to those skilled in the art. Such alterations, modifications, and improvements are intended to be within the spirit and scope of the technology described herein. For example, those of ordinary skill in the art will readily envision a variety of other means and/or structures for performing the function and/or obtaining the results and/or one or more of the advantages described herein, and each of such variations and/or modifications is deemed to be within the scope of the embodiments described herein. Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many equivalents to the specific embodiments described herein. It is, therefore, to be understood that the foregoing embodiments are presented by way of example only and that, within the scope of the appended claims and equivalents thereto, inventive embodiments may be practiced otherwise than as specifically described. In addition, any combination of two or more features, systems, articles, materials, kits, and/or methods described herein, if such features, systems, articles, materials, kits, and/or methods are not mutually inconsistent, is included within the scope of the present disclosure.
The above-described embodiments can be implemented in any of numerous ways. One or more aspects and embodiments of the present disclosure involving the performance of processes or methods may utilize program instructions executable by a device (e.g., a computer, a processor, or other device) to perform, or control performance of, the processes or methods. In this respect, various inventive concepts may be embodied as a computer readable storage medium (or multiple computer readable storage media) (e.g., a computer memory, one or more floppy discs, compact discs, optical discs, magnetic tapes, flash memories, circuit configurations in Field Programmable Gate Arrays or other semiconductor devices, or other tangible computer storage medium) encoded with one or more programs that, when executed on one or more computers or other processors, perform methods that implement one or more of the various embodiments described above. The computer readable medium or media can be transportable, such that the program or programs stored thereon can be loaded onto one or more different computers or other processors to implement various ones of the aspects described above. In some embodiments, computer readable media may be non-transitory media.
The terms “program” or “software” are used herein in a generic sense to refer to any type of computer code or set of computer-executable instructions that can be employed to program a computer or other processor to implement various aspects as described above. Additionally, it should be appreciated that according to one aspect, one or more computer programs that when executed perform methods of the present disclosure need not reside on a single computer or processor, but may be distributed in a modular fashion among a number of different computers or processors to implement various aspects of the present disclosure.
Computer-executable instructions may be in many forms, such as program modules, executed by one or more computers or other devices. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. Typically the functionality of the program modules may be combined or distributed as desired in various embodiments.
Also, data structures may be stored in computer-readable media in any suitable form. For simplicity of illustration, data structures may be shown to have fields that are related through location in the data structure. Such relationships may likewise be achieved by assigning storage for the fields with locations in a computer-readable medium that convey relationship between the fields. However, any suitable mechanism may be used to establish a relationship between information in fields of a data structure, including through the use of pointers, tags or other mechanisms that establish relationship between data elements.
When implemented in software, the software code can be executed on any suitable processor or collection of processors, whether provided in a single computer or distributed among multiple computers.
Further, it should be appreciated that a computer may be embodied in any of a number of forms, such as a rack-mounted computer, a desktop computer, a laptop computer, or a tablet computer, as non-limiting examples. Additionally, a computer may be embedded in a device not generally regarded as a computer but with suitable processing capabilities, including a Personal Digital Assistant (PDA), a smartphone or any other suitable portable or fixed electronic device.
Also, a computer may have one or more input and output devices. These devices can be used, among other things, to present a user interface. Examples of output devices that can be used to provide a user interface include printers or display screens for visual presentation of output and speakers or other sound generating devices for audible presentation of output. Examples of input devices that can be used for a user interface include keyboards, and pointing devices, such as mice, touch pads, and digitizing tablets. As another example, a computer may receive input information through speech recognition or in other audible formats.
Such computers may be interconnected by one or more networks in any suitable form, including a local area network or a wide area network, such as an enterprise network, and intelligent network (IN) or the Internet. Such networks may be based on any suitable technology and may operate according to any suitable protocol and may include wireless networks, wired networks or fiber optic networks.
Also, as described, some aspects may be embodied as one or more methods. The acts performed as part of the method may be ordered in any suitable way. Accordingly, embodiments may be constructed in which acts are performed in an order different than illustrated, which may include performing some acts simultaneously, even though shown as sequential acts in illustrative embodiments.
All definitions, as defined and used herein, should be understood to control over dictionary definitions, definitions in documents incorporated by reference, and/or ordinary meanings of the defined terms.
The indefinite articles “a” and “an,” as used herein in the specification and in the claims, unless clearly indicated to the contrary, should be understood to mean “at least one.”
The phrase “and/or,” as used herein in the specification and in the claims, should be understood to mean “either or both” of the elements so conjoined, i.e., elements that are conjunctively present in some cases and disjunctively present in other cases. Multiple elements listed with “and/or” should be construed in the same fashion, i.e., “one or more” of the elements so conjoined. Other elements may optionally be present other than the elements specifically identified by the “and/or” clause, whether related or unrelated to those elements specifically identified. Thus, as a non-limiting example, a reference to “A and/or B”, when used in conjunction with open-ended language such as “comprising” can refer, in one embodiment, to A only (optionally including elements other than B); in another embodiment, to B only (optionally including elements other than A); in yet another embodiment, to both A and B (optionally including other elements); etc.
As used herein in the specification and in the claims, the phrase “at least one,” in reference to a list of one or more elements, should be understood to mean at least one element selected from any one or more of the elements in the list of elements, but not necessarily including at least one of each and every element specifically listed within the list of elements and not excluding any combinations of elements in the list of elements. This definition also allows that elements may optionally be present other than the elements specifically identified within the list of elements to which the phrase “at least one” refers, whether related or unrelated to those elements specifically identified. Thus, as a non-limiting example, “at least one of A and B” (or, equivalently, “at least one of A or B,” or, equivalently “at least one of A and/or B”) can refer, in one embodiment, to at least one, optionally including more than one, A, with no B present (and optionally including elements other than B); in another embodiment, to at least one, optionally including more than one, B, with no A present (and optionally including elements other than A); in yet another embodiment, to at least one, optionally including more than one, A, and at least one, optionally including more than one, B (and optionally including other elements); etc.
Also, the phraseology and terminology used herein is for the purpose of description and should not be regarded as limiting. The use of “including,” “comprising,” or “having,” “containing,” “involving,” and variations thereof herein, is meant to encompass the items listed thereafter and equivalents thereof as well as additional items.
In the claims, as well as in the specification above, all transitional phrases such as “comprising,” “including,” “carrying,” “having,” “containing,” “involving,” “holding,” “composed of,” and the like are to be understood to be open-ended, i.e., to mean including but not limited to. Only the transitional phrases “consisting of” and “consisting essentially of” shall be closed or semi-closed transitional phrases, respectively.
Ralston, Tyler S., Sanchez, Nevada J.
Patent | Priority | Assignee | Title |
11815492, | Apr 16 2020 | BFLY OPERATIONS, INC. | Methods and circuitry for built-in self-testing of circuitry and/or transducers in ultrasound devices |
Patent | Priority | Assignee | Title |
6200266, | Mar 31 1998 | Case Western Reserve University | Method and apparatus for ultrasound imaging using acoustic impedance reconstruction |
7549961, | Jul 31 2003 | FUJIFILM SONOSITE, INC | System and method supporting imaging and monitoring applications |
8277380, | Sep 11 2007 | Siemens Medical Solutions USA, Inc. | Piezoelectric and CMUT layered ultrasound transducer array |
8647279, | Jun 10 2010 | Siemens Medical Solutions USA, Inc. | Volume mechanical transducer for medical diagnostic ultrasound |
8852103, | Oct 17 2011 | BFLY OPERATIONS, INC | Transmissive imaging and related apparatus and methods |
9229097, | Apr 18 2014 | BFLY OPERATIONS, INC | Architecture of single substrate ultrasonic imaging devices, related apparatuses, and methods |
9521991, | Mar 15 2013 | BFLY OPERATIONS, INC | Monolithic ultrasonic imaging devices, systems and methods |
9592030, | Jul 23 2013 | BFLY OPERATIONS, INC | Interconnectable ultrasound transducer probes and related methods and apparatus |
9592032, | Apr 18 2014 | BFLY OPERATIONS, INC | Ultrasonic imaging compression methods and apparatus |
20030028113, | |||
20030097071, | |||
20070085606, | |||
20070242567, | |||
20080092657, | |||
20080114253, | |||
20130172752, | |||
20140050048, | |||
20140064022, | |||
20140180094, | |||
20140204700, | |||
20140269166, | |||
20160199030, | |||
20160331353, | |||
20170307741, | |||
20190090854, | |||
EP83985, | |||
TW201325556, | |||
WO2009135255, |
Executed on | Assignor | Assignee | Conveyance | Frame | Reel | Doc |
Oct 07 2014 | RALSTON, TYLER S | Butterfly Network, Inc | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 048139 | /0639 | |
Oct 07 2014 | SANCHEZ, NEVADA J | Butterfly Network, Inc | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 048139 | /0639 | |
Jan 18 2019 | Butterfly Network, Inc. | (assignment on the face of the patent) | / | |||
Feb 12 2021 | Butterfly Network, Inc | BFLY OPERATIONS, INC | CHANGE OF NAME SEE DOCUMENT FOR DETAILS | 058823 | /0668 |
Date | Maintenance Fee Events |
Jan 18 2019 | BIG: Entity status set to Undiscounted (note the period is included in the code). |
Feb 08 2019 | SMAL: Entity status set to Small. |
Feb 10 2022 | BIG: Entity status set to Undiscounted (note the period is included in the code). |
Feb 06 2023 | M1551: Payment of Maintenance Fee, 4th Year, Large Entity. |
Date | Maintenance Schedule |
Aug 06 2022 | 4 years fee payment window open |
Feb 06 2023 | 6 months grace period start (w surcharge) |
Aug 06 2023 | patent expiry (for year 4) |
Aug 06 2025 | 2 years to revive unintentionally abandoned end. (for year 4) |
Aug 06 2026 | 8 years fee payment window open |
Feb 06 2027 | 6 months grace period start (w surcharge) |
Aug 06 2027 | patent expiry (for year 8) |
Aug 06 2029 | 2 years to revive unintentionally abandoned end. (for year 8) |
Aug 06 2030 | 12 years fee payment window open |
Feb 06 2031 | 6 months grace period start (w surcharge) |
Aug 06 2031 | patent expiry (for year 12) |
Aug 06 2033 | 2 years to revive unintentionally abandoned end. (for year 12) |